Trifid M20 (First Light SX-46)

[PRejto]

After months of struggling with this new camera finally the host of issues are resolved. The camera is rock solid now but the process took 4 mother board revisions, a linear external power supply, and resolution of an incompatibility with the Mount Hub Pro-compact due to a grounding issue. In any case I guess this is what one gets as an early adopter. Anyone contemplating the camera can be sure that it works as advertised (now) but beware that you need a very clean power supply (my switching supplies would not operate the camera) and the USB is very sensitive to cooling demands and will try to draw power however it may. The MHP-c grounds in a normal way but the SX-46 does not like this (my Moravian camera had no issue, but the power current requirements are a lot less). I now use the MHP-c to power an external relay to control the camera. This solved the grounding issue. Images are quite clean and download rapidly.


EDIT: I have been remiss in forgetting to mention that my equipment is located at Heaven's Mirror Observatory in Yass. Martin Pugh has been an invaluable asset in helping me with numerous technical issues both optical and with the SX-46!



This is a straight RGB image of about 3 hours each channel. (TEC180FL + flattener) I could find no benefit from the luminance I captured. I probably could benefit from more blue. The Astronomik DeepSky filters capture about 30% less light than the older blue Astronomic Type 2c filters I'm more used to in my previous camera.


This is a tough object to stretch. After many failed attempts I finally got the masked stretch in Pix to operate without making crazy looking stars. I finished the stretch with the HT tool. I wish I could have gotten more blue into the fainter regions.....


Full version is on Astrobin which for some reason changes the feel of the image in a way I don't appreciate. In Pix and Photoshop there is more "warmth." I uploaded a full resolution png; the initial result looks too black in background and background star colour too flat. The 2nd version I uploaded is better but still lacking compared to what I see on my monitor before uploading. Does anyone else have this issue with Astrobin?



Thanks for looking!


New version: https://www.astrobin.com/full/369759/G/


original: https://www.astrobin.com/full/369759/0/ (https://www.astrobin.com/full/369759/0/)

[Imme]

Very nice image. One of the best I've seen on here I think

[Andy01]

Nice one Peter, well persevered!
A good Triffid never gets old, well done :thumbsup:

[RickS]

Hi Peter,

My commiserations on all the issues you have faced but the final outcome is impressive!

I downloaded a copy of the image from Astrobin and it doesn't contain an ICC colour profile. That may be the cause of some of your colour inconsistencies. Astrobin used to strip ICC Profiles but that was fixed some time ago. IIRC, PixInsight doesn't know how to embed an ICC profile in a PNG file so if you want to use PNG format you might need to fix that with Photoshop.

Cheers,
Rick.

[PRejto]

Thanks Jon, Andy and Rick for the kind words!


Rick, I actually saved the png from CS-5. I just checked the profile and it is sRGB IEC61966-2.1 on the image. I can reassign it to Adobe RGB, or other if you think that would prove useful. Since a .tif cannot be uploaded to Astrobin I assumed (incorrectly?) that a png file would be better than any jpg, no? What do you upload?


Should I reassign to AdobeRGB and reup?



Thanks,


Peter

[RickS]

Hi Peter,

In that case the profile is getting stripped off somewhere between your upload and my download. I just did a quick test and the same thing happened, so perhaps Astrobin is broken again. I'll see if I can find out...

I normally upload a reasonable quality JPEG file and keep the lossless versions for printing :)

These days I suspect that AdobeRGB is a better lowest common denominator profile than sRGB. Even cheap LCD screens have a reasonably wide gamut.

Cheers,
Rick.

[willik]

Very nice triffid good colour and detail
Martin

[Atmos]

Great Triffid Peter! Great resolution, colour, depth and contrast :thumbsup:

[cometcatcher]

Wow, that's a beauty!

[Ryderscope]

A fine image Peter and an interesting discussion on the issues with the ICC profiles. I’ve been looking at this lately myself with my Astrobin posts as well so it’s good to track theses discussions. You’ve done well with the RGB data here showing some nice distribution of neubulosity across the image.

[PRejto]

Thanks Rodney, Kevin, Colin, and Martin! Your comments are much appreciated.


I re-upped a Adobe RGB1998 and it has a completely different look compared to opening the same image in Pix or CS-5. I downloaded the jpg from Astrobin (which shows same Adobe RGB profile)and opened it in CS-5 and it is VERY comparable to the image I uploaded. Then I opened the image with a different windows view program (ImageGlass) and it looks just like the browser displayed image (Firefox). So is this some sort of Win10 issue? Is there an adjustment in windows to correct this?


Thanks!
Peter

[RickS]

A lot of applications (including browsers) either don't do colour management or don't do it correctly and this isn't just restricted to Windows 10.

Here's a page with some useful test images: http://www.color.org/version4html.xalter

Cheers,
Rick.

[PRejto]

Thanks for that Rick!



I saved the 4 images and they all look normal on my machine. So, what is different when say CS-5 displays an image compared to when a browser displays an image? What is making them appear different?


Here is a little comparison....a screen cap of the same image side by side. one in Photoshop, the other ImageGlass (which looks very much like the image displays in Firefox). The color is more muted and background is odd.

[RickS]

I'm fairly sure that Photoshop uses an Adobe developed colour management system. The browsers are probably using the Windows API or some other less well developed implementation. Some older versions of browsers don't do colour management at all and this varies between platforms as well.

Take a look at this: https://www.color-management-guide.com/web-browser-color-management.html There's a note in there about how to change Firefox settings from the default to get better colour management.

Cheers,
Rick.

[PRejto]

I guess this begs the question of how we can ever be sure about how another person evaluates our images in spite of our best efforts to calibrate!


Thanks for the link!


Peter

[RickS]

Alas, I think it's a lot like walking into an electronics store and watching the wall of TVs all showing the same program in wildly different colours.

[Slawomir]

Fantastic image Peter and well done on persevering with initial issues - looks like the camera works very well now and will provide great data :thumbsup:

[PRejto]

Thanks Suavi!



If only the camera had worked as it should have from day 1! I guess this hobby makes all of us beta testers! How rare is it that anything goes as expected?


But, I will say this, Terry Platt gives phenomenal personal attention to every problem.



Peter

[PRejto]

Rick,



The article you quoted was very helpful with the recommendation to save every internet image as sRGB. I upped a new sRGB to Astrobin and the improvement is quite striking. What formerly appeared (to me from Astrobin through Firefox) looked a bit desaturated but is now pretty normal. One can only guess what anybody else actually sees!


Thanks again,


Peter

[RickS]

I'm glad it was helpful, Peter. Images in a wide gamut colour space like AdobeRGB or ProPhotoRGB will look drab if (mis)interpreted as sRGB so I guess it is still the safe choice.

Cheers,
Rick.

[strongmanmike]

Quite a nice Trifid Peter, good to hear you have the SX camera sorted too. I have found my SX gear to be pretty good all these years, so hopefully your SX46 will now give years of trouble free performance :thumbsup:

My wife and I were out again on the weekend looking for our own suitable rural property to hopefully do something similar... and I finally found Martins house and site. Drove past it on Yass Valley Way and recognised it from the road given the snipets of clues available in his website photos. I had tried looking for it on Google maps before but could never identify it. Looks like a good spot.

Mike

[Paul Haese]

The link is showing a fata error message Peter, but the thumbnail looks nice. I'll look again once that link is working.

[gregbradley]

Hi Peter,

Nice to see an image from your setup.
Nice background and stars in the image but the colour of the Trifid is a bit off. You've got an odd shade of pink/magenta in the coloured neb and the blue reflection nebula colour doesn't seem the correct shade as well.

Perhaps look at Rob Gendler's version as a benchmark. He to my eye, has the best colour management. Another would be Adam Block. Its probably your Ha blend has shifted the colour of the main neb too much.

Greg.

[PRejto]

Thanks Mike!


I thought giving up my multi-camera system would make life easier. Ha, I seriously underestimated how difficult remote imaging can be. I don't even want to count up the number of times I've driven to Yass from Sydney! Good luck with your observatory project.


Peter

[PRejto]

Hi Paul,


Thanks for letting me know about the link. Strangely, it works for me, but in checking it was clearly bringing up the wrong image. So, I've deleted everything except the latest revision and now it seems to be working OK. I would appreciate your take on my efforts here.


Many thank,
Peter

[PRejto]

Hi Greg,


I appreciate your comments! It's possible that you were looking at an image that I intended not to show...nevertheless, you might still feel the same about the current version. I did look at Gendler's just now and I agree about the red. I might try and tweak mine but I have to say that I'm a bit past trying to start over yet again. By the way, I did not use any Ha data...only RGB, no luminance either.


Edit: attached an attempt on the magenta. Is this going in the right direction?



All best,
Peter

[PRejto]

I've uploaded a new version that I hope has improved colour. (see first post) I also was able to dig out some additional detail using the APF-R method (with a mask).


I'd appreciate any comments. Hopefully I'm moving in the right direction!


Thanks,


Peter

[Googaliser]

Terrific image !

[gregbradley]

Yes I think that looks better. The blue reflection neb is a bit more restrained and showing the dimmer outer bands which are not as blue and the neb colour seems good. No Ha eh? Wow, you picked it up anyway with the other filters.

That fluorite lens eh? Fluorite lensed scopes do seem to produce better colours.

The lovely starry background is particularly good.

Greg.

[PRejto]

Hi Greg,


Thanks for your message. I have tried and tried but I cannot bring out more blue in the fainter areas. Maybe there is some trick in processing but I can't find it. But, I do think I need more blue frames...perhaps 30% more than the other filters based on some preliminary tests with the Deep Sky set.


Thanks,


Peter